期刊,Robotics & Machine Learning Daily News 2024年卷Mar.8期_国家学术搜索

期刊信息/Journal information

Robotics & Machine Learning Daily News

NewsRx

主办单位：NewsRx

Robotics & Machine Learning Daily News/Journal Robotics & Machine Learning Daily News

正式出版

收录年代

Findings in Artificial Intelligence Reported from Zhejiang University (Artificia l-intelligence-based Hybrid Extended Phase Shift Modulation for the Dual Active Bridge Converter With Full Zvs Range and Optimal Efficiency)

10-11页

查看更多>>摘要：By a News Reporter-Staff News Editor at Robotics & Machine Learning Daily News Daily News-Fresh data on Artificial Intelligence are presented in a new report. According to news originating from Hangzhou, Peop le's Republic of China, by NewsRx correspondents, research stated, "The dual act ive bridge (DAB) converter is the key enabler in many popular applications, such as wireless charging, electric vehicle, and renewable energy. ZVS range and eff iciency are two significant performance indicators for the DAB converter." Our news journalists obtained a quote from the research from Zhejiang University, "To obtain the desired ZVS and efficiency performance, modulation should be ca refully designed. Hybrid modulation (HM) considers several single modulation str ategies to achieve good comprehensive performance. Conventionally, to design an HM, a harmonic approach or piecewise approach is used, but they suffer from a ti me-consuming model-building process and inaccuracy. Therefore, an artificial-int elligence-based hybrid extended phase shift (HEPS) modulation is proposed. Gener ally, the HEPS modulation is developed in an automated fashion, which alleviates the cumbersome model-building process while keeping high model accuracy. In HEP S modulation, two EPS strategies are considered to realize optimal efficiency wi th full ZVS operation over entire operating ranges. Specifically, to build data- driven models of ZVS and efficiency performance, extreme gradient boosting (XGBo ost), which is a state-of-the-art ensemble learning algorithm, is adopted. After ward, particle swarm optimization with state-based adaptive velocity limit (PSO- SAVL) is utilized to select the best EPS strategy and optimize modulation parame ters."

原文链接:

NETL
NSTL

New Robotics Findings from Zhejiang University Discussed (Modeling and Mpc-based Pose Tracking for Wheeled Bipedal Robot)

11-12页

查看更多>>摘要：By a News Reporter-Staff News Editor at Robotics & Machine Learning Daily News Daily News-Researchers detail new data in Robotic s. According to news reporting from Hangzhou, People's Republic of China, by New sRx journalists, research stated, "In this letter, we propose a model predictive control (MPC)-based robot pose controller for our newly designed wheeled bipeda l robot (WBR). The proposed controller uses the virtual model control concept, a llowing for wider applicability by ignoring the leg dynamics." Financial support for this research came from National Natural Science Foundatio n of China (NSFC). The news correspondents obtained a quote from the research from Zhejiang Univers ity, "By directly incorporating the non-holonomic constraint of the wheels into the dynamic equation, a wheeled rigid dynamic model is proposed to maximize the motion flexibility and minimize the model order. A hierarchical MPC control stru cture is employed to track the desired pose while considering the non-minimal ph ase property of WBRs in real time. To enhance the autonomy of the robot, we prop ose a state estimator that utilizes kinematics and inertial sensor data to provi de a high-speed and accurate estimation of the robot's state. Both simulation an d real-world experiments demonstrate that the proposed method can track a pose t rajectory with lower error than traditional feedback control methods."

原文链接:

NETL
NSTL

Researchers from University of Ottawa Report Recent Findings in Robotics and Aut omation (Stability of Human Balance During Quiet Stance With Physiological and E xoskeleton Time Delays)

12-12页

查看更多>>摘要：By a News Reporter-Staff News Editor at Robotics & Machine Learning Daily News Daily News-Data detailed on Robotics - Robotics a nd Automation have been presented. According to news reporting originating from Ottawa, Canada, by NewsRx correspondents, research stated, "Human balance with e xoskeleton assistance is studied using an inverted pendulum model, considering t ime delays in the muscle reflexes and the exoskeleton controller. The model incl udes two motors at the ankle joint whose maximum torques depend on the joint ang le and angular velocity, reflecting the combined moment-generating capacity of a ll plantarflexor and dorsiflexor muscles." Financial support for this research came from CGIAR. Our news editors obtained a quote from the research from the University of Ottaw a, "These ‘musclelike' motors obey a proportional-derivative (PD) reflex contro l law where the angle and angular velocity of the ankle joint are subject to fee dback delays. The stability of this system is analyzed using Galerkin projection to convert the governing neutral delay differential equation into a system of f irst-order ordinary differential equations (ODEs) and computing the eigenvalues of the ODE system. The stability analysis is then repeated with exoskeleton torq ues included at the ankle joint. The exoskeleton torques are assumed to obey a P D control law as well but with a unique state feedback delay. Stability charts r eveal that the area of the stability region always increases as the exoskeleton delay decreases, but the area may decrease as the physiological delay decreases. "

原文链接:

NETL
NSTL

Shanghai Jiao Tong University Reports Findings in Machine Learning (Prediction o f chemical reaction yields with large-scale multiview pre-training)

13-13页

查看更多>>摘要：By a News Reporter-Staff News Editor at Robotics & Machine Learning Daily News Daily News-New research on Machine Learning is th e subject of a report. According to news reporting originating from Shanghai, Pe ople's Republic of China, by NewsRx correspondents, research stated, "Developing machine learning models with high generalization capability for predicting chem ical reaction yields is of significant interest and importance. The efficacy of such models depends heavily on the representation of chemical reactions, which h as commonly been learned from SMILES or graphs of molecules using deep neural ne tworks." Financial support for this research came from National Natural Science Foundatio n of China. Our news editors obtained a quote from the research from Shanghai Jiao Tong Univ ersity, "However, the progression of chemical reactions is inherently determined by the molecular 3D geometric properties, which have been recently highlighted as crucial features in accurately predicting molecular properties and chemical r eactions. Additionally, large-scale pre-training has been shown to be essential in enhancing the generalization capability of complex deep learning models. Base d on these considerations, we propose the Reaction Multi-View Pre-training (ReaM VP) framework, which leverages self-supervised learning techniques and a two-sta ge pre-training strategy to predict chemical reaction yields. By incorporating m ulti-view learning with 3D geometric information, ReaMVP achieves state-of-the-a rt performance on two benchmark datasets. Notably, the experimental results indi cate that ReaMVP has a significant advantage in predicting out-of-sample data, s uggesting an enhanced generalization ability to predict new reactions. Scientifi c Contribution: This study presents the ReaMVP framework, which improves the gen eralization capability of machine learning models for predicting chemical reacti on yields. By integrating sequential and geometric views and leveraging self-sup ervised learning techniques with a two-stage pre-training strategy, ReaMVP achie ves state-of-the-art performance on benchmark datasets."

原文链接:

NETL
NSTL

Aston University Reports Findings in Artificial Intelligence (Utility of artific ial intelligence-based large language models in ophthalmic care)

14-15页

查看更多>>摘要：By a News Reporter-Staff News Editor at Robotics & Machine Learning Daily News Daily News-New research on Artificial Intelligenc e is the subject of a report. According to news reporting from Birmingham, Unite d Kingdom, by NewsRx journalists, research stated, "With the introduction of Cha tGPT, artificial intelligence (AI)-based large language models (LLMs) are rapidl y becoming popular within the scientific community. They use natural language pr ocessing to generate human-like responses to queries." The news correspondents obtained a quote from the research from Aston University, "However, the application of LLMs and comparison of the abilities among differ ent LLMs with their human counterparts in ophthalmic care remain under-reported. Hitherto, studies in eye care have demonstrated the utility of ChatGPT in gener ating patient information, clinical diagnosis and passing ophthalmology question -based examinations, among others. LLMs' performance (median accuracy, % ) is influenced by factors such as the iteration, prompts utilised and the domai n. Human expert (86%) demonstrated the highest proficiency in disea se diagnosis, while ChatGPT-4 outperformed others in ophthalmology examinations (75.9%), symptom triaging (98%) and providing informat ion and answering questions (84.6%). LLMs exhibited superior perfor mance in general ophthalmology but reduced accuracy in ophthalmic subspecialties . Although AI-based LLMs like ChatGPT are deemed more efficient than their human counterparts, these AIs are constrained by their nonspecific and outdated train ing, no access to current knowledge, generation of plausible-sounding ‘fake' res ponses or hallucinations, inability to process images, lack of critical literatu re analysis and ethical and copyright issues. A comprehensive evaluation of rece ntly published studies is crucial to deepen understanding of LLMs and the potent ial of these AI-based LLMs. Ophthalmic care professionals should undertake a con servative approach when using AI, as human judgement remains essential for clini cal decision-making and monitoring the accuracy of information. This review iden tified the ophthalmic applications and potential usages which need further explo ration. With the advancement of LLMs, setting standards for benchmarking and pro moting best practices is crucial."

原文链接:

NETL
NSTL

China University of Petroleum Researcher Provides New Insights into Machine Lear ning (Anthropomorphic Soft Hand: Dexterity, Sensing, and Machine Learning)

15-15页

查看更多>>摘要：By a News Reporter-Staff News Editor at Robotics & Machine Learning Daily News Daily News-Fresh data on artificial intelligence are presented in a new report. According to news originating from Beijing, Peopl e's Republic of China, by NewsRx correspondents, research stated, "Humans posses s dexterous hands that surpass those of other animals, enabling them to perform intricate, complex movements." Financial supporters for this research include Beijing Natural Science Foundatio n. Our news correspondents obtained a quote from the research from China University of Petroleum: "Soft hands, known for their inherent flexibility, aim to replica te the functionality of human hands. This article provides an overview of the de velopment processes and key directions in soft hand evolution. Starting from bas ic multi-finger grippers, these hands have made significant advancements in the field of robotics. By mimicking the shape, structure, and functionality of human hands, soft hands can partially replicate human-like movements, offering adapta bility and operability during grasping tasks. In addition to mimicking human han d structure, advancements in flexible sensor technology enable soft hands to exh ibit touch and perceptual capabilities similar to humans, enhancing their perfor mance in complex tasks. Furthermore, integrating machine learning techniques has significantly promoted the advancement of soft hands, making it possible for th em to intelligently adapt to a variety of environments and tasks." According to the news reporters, the research concluded: "It is anticipated that these soft hands, designed to mimic human dexterity, will become a focal point in robotic hand development. They hold significant application potential for ind ustrial flexible gripping solutions, medical rehabilitation, household services, and other domains, offering broad market prospects."

原文链接:

NETL
NSTL

Findings from Dalian University of Technology Broaden Understanding of Robotics and Automation (Vi-hso: Hybrid Sparse Monocular Visual-inertial Odometry)

16-17页

查看更多>>摘要：By a News Reporter-Staff News Editor at Robotics & Machine Learning Daily News Daily News-Current study results on Robotics - Ro botics and Automation have been published. According to news reporting originati ng in Dalian, People's Republic of China, by NewsRx journalists, research stated, "In this letter, we present VI-HSO, a hybrid sparse monocular visual-inertial odometry system based on two innovative techniques called adaptive interframe al ignment (AIA) and dynamic inverse distance filter (DIDF). Although the sparse im age alignment algorithm appears efficient for calculating frame-to-frame motion, it tends to fail in case of significant intensity changes and motion blur." Financial supporters for this research include National Natural Science Foundati on of China (NSFC), Shenzhen Key Laboratory of Robotics and Computer Vision.

原文链接:

NETL
NSTL

Reports Summarize Machine Learning Research from Catholic University of the Sacr ed Heart (Toward Greener Smart Cities: A Critical Review of Classic and Machine- Learning-Based Algorithms for Smart Bin Collection)

16-16页

查看更多>>摘要：By a News Reporter-Staff News Editor at Robotics & Machine Learning Daily News-Current study results on artificial intelligence have been published. According to news originating from Brescia, Italy, by NewsR x correspondents, research stated, "This study critically reviews the scientific literature regarding machine-learning approaches for optimizing smart bin colle ction in urban environments." Our news correspondents obtained a quote from the research from Catholic Univers ity of the Sacred Heart: "Usually, the problem is modeled within a dynamic graph framework, where each smart bin's changing waste level is represented as a node . Algorithms incorporating Reinforcement Learning (RL), time-series forecasting, and Genetic Algorithms (GA) alongside Graph Neural Networks (GNNs) are analyzed to enhance collection efficiency." According to the news reporters, the research concluded: "While individual metho dologies present limitations in computational demand and adaptability, their syn ergistic application offers a holistic solution. From a theoretical point of vie w, we expect that the GNN-RL model dynamically adapts to real-time data, the GNN -time series predicts future bin statuses, and the GNN-GA hybrid optimizes netwo rk configurations for accurate predictions, collectively enhancing waste managem ent efficiency in smart cities."

原文链接:

NETL
NSTL

University of Belgrade Reports Findings in Machine Learning (Prefrontal cortical synaptoproteome profile combined with machine learning predicts resilience towa rds chronic social isolation in rats)

18-19页

查看更多>>摘要：By a News Reporter-Staff News Editor at Robotics & Machine Learning Daily News-New research on Machine Learning is the subject o f a report. According to news reporting out of Belgrade, Serbia, by NewsRx edito rs, research stated, "Chronic social isolation (CSIS) of rats serves as an anima l model of depression and generates CSIS-resilient and CSIS-susceptible phenotyp es. We aimed to investigate the prefrontal cortical synaptoproteome profile of C SIS-resilient, CSIS-susceptible, and control rats to delineate biochemical pathw ays and predictive biomarker proteins characteristic for the resilient phenotype ." Our news journalists obtained a quote from the research from the University of B elgrade, "A sucrose preference test was performed to distinguish rat phenotypes. Class separation and machine learning (ML) algorithms support vector machine wi th greedy forward search and random forest were then used for discriminating CSI S-resilient from CSIS-susceptible and control rats. CSIS-resilient compared to C SIS-susceptible rat proteome analysis revealed, among other proteins, downregula ted glycolysis in- termediate fructose-bisphosphate aldolase C (Aldoc), and upregu lated clathrin heavy chain 1 (Cltc), calcium/calmodulin-dependent protein kinase type II (Cam2a), synaptophysin (Syp) and fatty acid synthase (Fasn) that are in volved in neuronal transmission, synaptic vesicular trafficking, and fatty acid synthesis. Comparison of CSIS-resilient and control rats identified downregulate d mitochondrial proteins ATP synthase subunit beta (Atp5f1b) and citrate synthas e (Cs), and upregulated protein kinase C gamma type (Prkcg), vesicular glutamate transporter 1 (Slc17a7), and synaptic vesicle glycoprotein 2 A (Sv2a) involved in signal transduction and synaptic trafficking. The combined protein difference s make the rat groups linearly separable, and 100% validation accu racy is achieved by standard ML models. ML algorithms resulted in four panels of discriminative proteins."

原文链接:

NETL
NSTL

Researcher from Qatar University Publishes New Studies and Findings in the Area of Artificial Intelligence (AI-Powered Innovation in Digital Transformation: Key Pillars and Industry Impact)

18-18页

查看更多>>摘要：By a News Reporter-Staff News Editor at Robotics & Machine Learning Daily News Daily News-Investigators publish new report on ar tificial intelligence. According to news originating from Qatar University by Ne wsRx editors, the research stated, "Digital transformation systems generate a su bstantial volume of data, creating opportunities for potential innovation, parti cularly those driven by artificial intelligence." Funders for this research include Qatar National Research Fund; Qatar National L ibrary. The news editors obtained a quote from the research from Qatar University: "This study focuses on the intricate relationship between artificial intelligence and innovation as foundational elements in the digital transformation framework for sustained growth and operational excellence. This study provides a holistic per spective on the cultivation and pillars of AI-powered innovation, highlighting t heir pivotal role in revolutionizing industries, including healthcare, education, finance, manufacturing, transportation, and agriculture. The work emphasizes t he key pillars essential for fostering AI-powered innovation, including monitori ng performance measurement to use the power of the present, continuous learning and innovation, data analytics and insights, predictive analytics, and innovativ e product development. This study investigates how these pillars serve as the fo undation for groundbreaking advancements, driving efficiency, enhancing decision -making processes, and fostering creativity within organizations."

原文链接:

NETL
NSTL